Equilibrium-Line Altitudes of the Present and Last Glacial Maximum in the Eastern Nepal Himalayas and Their Implications for SW Monsoon Climate

Quaternary International 212 (2010) 26–34

Contents lists available at ScienceDirect

Quaternary International

journal homepage: www.elsevier.com/locate/quaint

Equilibrium-line altitudes of the present and Last Glacial Maximum in the eastern Nepal Himalayas and their implications for SW monsoon climate

Katsuhiko Asahi*

Faculty of Environmental Earth Science, Hokkaido University, Sapporo 060-0810, Japan article info abstract

Article history: Equilibrium-line altitude (ELA) of a glacier can be an adequate indicator for glaciation and local climate. Available online 18 September 2008 This study addresses the present ELA in eastern Nepal, in comparison with that of the Last Glacial Maximum (LGM). The present ELA of each glacier was identified by its surface topography using the latest aerial photograph interpretations. ELAs during the LGM were mostly estimated by the maximum elevation of lateral moraines (MELMs). The distinguished higher altitudes of the MELM were selected to estimate the ELAs during the LGM. Latitudinal profiles of the present equilibrium-line show southward gradients, interpreted as a product of the reduction in precipitation from monsoon humid wind blowing into the High Himalayas. Latitudinal ELA profiles of the LGM show the same southward inclination, suggesting that monsoon precipitation is likely to be a ruling resource on glacier nourishment. The eastern Nepal Himalayas, hence, was under summer monsoon environment during the LGM same as present, even if the SW monsoons were weaker. Ó 2008 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction rudimentary problems in the Himalayas, those being: (1) the lack of numerical dating means that the chronologies of the glaciation The Himalayas are characterized by the existence of glaciers at have not been adequately understood, (2) the inherent errors of the high altitudes. During the Last Glacial, glaciers covered a vastly ELA reconstructing method due to the orographic steepness of the greater area than at present, and many of the high altitude regions Himalayas, and (3) the effect of the latitudinal inclination of ELA on in the Nepal Himalayas have been glacierized. Glacial landforms the accuracy of the calculation of DELA. indicate the existence of past glaciation, and the limits of glacier In the last decade, numerical dating methods for glacial sedi- extent may suggest paleoclimatic conditions. The Nepal Himalayas ments have been developed, and several studies on glacial history are located under the influence of SW summer monsoon environ- were conducted in the Himalayas. Recently, these techniques have ments, and monsoon precipitation plays an important role in also been applied to the Nepal Himalayas. From east to west, glacier accumulation and fluctuation. Hence, study of previous terrestrial cosmogenic radionuclides (TCR) or optically stimulated glaciation can reveal monsoon fluctuations in the Himalayas. luminescence (OSL) age has been designated in the Kanchenjunga Glacial equilibrium-line altitude (ELA) is the elevation where the (Tsukamoto et al., 2002), Khumbu (Richards et al., 2000; Finkel accumulation and ablation of a glacier are balanced. This is gener- et al., 2003), Langtang (Barnard et al., 2006), Ganesh (Abramowski ally regarded as the snowline altitude on glaciers and therefore, it is et al., 2006; Gayer et al., 2006) and Gorkha (Zech et al., 2003) an appropriate indicator for glaciation and local climate. In places Himals (Fig. 1), and application could successfully start to ascertain where modern glaciers exist, the lowering altitude of ELA (DELA) the timing of the major glacial advance during the Last Glacial can be used as a high-resolution proxy of air temperature and Maximum (LGM). TCR and OSL dates in conjunction with precipitation changes. Paleoclimatic reconstruction using the geomorphic and sedimentological analyses for selected regions of lowering altitude of past ELA, however, has been faced with the Himalayas would enable the determination of the extent of glaciation during the LGM (Richards, 2000). Further regional mapping and numerical dating are required to reconstruct and refine the pattern of glaciation of all high mountains of Central Asia. * Present address: Department of Geography, Ritsumeikan University, Kita-ku, Kyoto 603-8577, Japan. Tel.: þ81 75 465 1957; fax: þ81 75 465 8376. Moreover, it is strongly believed that detailed mapping of the E-mail address: [email protected] glacial landforms in the entire study area is essential not only to

Fig. 1. Locations of the Himalayan massifs in Nepal and the region where DELA has been reported. The light-coloured area shows over 4000 m a.s.l. and dark-coloured area over 6000 m a.s.l.

elucidate the classification of the glacial advanced stage by mor- description and evaluation of the ELA calculating site are quite phostratigraphy but also to determine the appropriate sites at important for correlating regional differences of ELAs, especially in which to collect TCR or OSL samples. Only after the completion of the Nepal Himalayas. such systematic studies with respect to many more Himalayan This study addresses to the ELA of the present glaciers in eastern regions will it be possible to produce a comprehensive map of the Nepal, in comparison with that of the Last Glacial one, and aims to extent of glaciation in the Himalayas during the LGM and for other speculate on the climatic change during the LGM. periods. As noted by Owen et al. (1997), it is necessary to adopt the relative dating methods, and extensive field observations for 2. Study area and regional climate improving the study on past glaciation in the Himalayas. Initially, Wissmann (1959) demonstrated the past ELA The Himalayas and the Tibetan Plateau influence regional and throughout Central Asia. Subsequent studies on reconstruction global climate systems (Benn and Owen, 1998). The Himalayan were done in the Nepal Himalayas and proposed the lowering chain restricts the flow of air and moisture between these moun- altitude of ELA during the LGM. These studies mainly adopted tains regions. Orographic precipitation caused by the southernmost conventional methods to reconstruct former ELA, including the area in the Himalayas depletes the moisture carried by summer SW accumulation-area ratio (AAR) method in Khumbu (Williams, 1983; monsoons, preventing moisture penetration to more northerly Burbank and Cheng, 1991) and in Marshandi (Burbank et al., 2003), mountains (Pant and Kumar, 1997). The Nepal Himalayas are the toe-to-headwall altitude ratio (THAR) method in the Narayani located at the central part of the Himalayas stretching east–west- watershed (Duncan et al., 1998) and in Gorkha (Zech et al., 2003), ward between 80 and 88 E(Fig. 1). The onset of the SW monsoon and the maximum elevation of lateral moraine (MELM) method in starts westward along the Himalayas from the north of the Bay of Khumbu (Mu¨ ller, 1980; Richards et al., 2000; Owen and Benn, Bengal and breaks up eastward. The Nepal Himalayas, especially in 2005), along with the empirical topographic model (Kuhle, 1988a) eastern regions of Nepal, is located under the strong SW monsoon in Kanchenjunga (Kuhle, 1990; Ko¨nig, 1999; Meiners, 1999) and environment, and monsoon precipitation plays an important role in Annapurna–Dhaulagiri (Kuhle, 1982) and the steady-state glacier glacier nourishment and fluctuation. Thus, the past monsoon mass-balance model in Langtang (Shiraiwa, 1993)(Fig. 1). However, environments can best be reconstructed in this area of Nepal. it is difficult to accurately estimate the past ELA using these The eastern Nepal Himalayas forms a high mountain chain along methods in the Himalayas. Benn and Lehmkuhl (2000) discussed the northernmost border. Among the 16 peaks in the world this problem comprehensively and mentioned that distinctive exceeding 8000 m, five peaks are located in this area. The modern characteristics of glacier mass-balance made determination of ELA glaciers extend linearly along the Great Himalaya range, and they difficult for both present and past due to the impact by avalanche- cover 1386 km2 (Asahi, 1999). Two areas were chosen in this study: fed, debris coverage, and orographic steepness of glaciers. There- the Khumbu and Kanchenjunga Himals. fore, there are inherent errors in calculation of past ELA for The Khumbu Himal includes Mt. Everest and other peaks higher a particular glacier with little geomorphic indication, which implies than 8000 m a.s.l. Numerous glaciers are located at heights past mass-balance conditions, especially in the Nepal Himalayas. approximately greater than 5000 m a.s.l. and the total glaciated Furthermore, if the regional equilibrium-line (EL) was inclined, area is 340 km2. Throughout this area, evidence of past glaciations the value of DELA differs depending on its place. A steep rise of is apparent from well-developed moraines and glacial deposits. The 30 m km1 from south to north was delineated in the Khumbu distribution of the glacial landforms and timing of past glaciations Himal (Mu¨ ller, 1980). The lowering altitude of past ELA can be along the Dudh Koshi valley have been argued since the 1970s, and arbitrarily variable depending on the points along its latitudinal these studies are reviewed by Owen et al. (1998). The arguments profile. Porter (2001) and Benn et al. (2005) emphasized that the were ascribed to different hypotheses in terms of determination of trend surface of ELAs should be calculated in order to ascertain the glacial landforms, namely the lower limit of glacial advance during most reliable estimate of DELA. Hence, the compilation of the LGM. Iwata (1976) supposed constrained glaciation, in which geomorphic mapping throughout the region as well as the the Khumbu Glacier advanced 6 km beyond the present terminus, 28 K. Asahi / Quaternary International 212 (2010) 26–34 down to 4200 m nearby Periche. In contrast, Fushimi (1978) sug- Tamur–Ghunsa Khola confluence valley. In contrast, Asahi and gested huge glacial extension which covered almost the entire Watanabe (2000) drew a geomorphological map throughout the Khumbu region, and the Khumbu Glacier extended ca. 40 km down Ghunsa Khola valley, and concluded that the maximum glacial to 2200 m near Lukla. Numerous studies followed either Iwata’s or extension was constrained down to Gyable at 2700 m a.s.l. Fushimi’s idea. One of the reasons for which alternative discussions Tsukamoto et al. (2002) assigned ages using OSL methods, and have not been settled is the lack of a detailed glacial landform map glacial advances were constrained at least during MIS 3, the LGM, of the entire area. Another major reason is the deficient numerical the Late Glacial, and the Holocene period. dating control on the timing of the glaciations except the Little Ice These numerical dating studies are surely epoch-making works Age. However, recently, several studies attempted to apply newly since these successfully elucidated the timings of glaciations, developed dating techniques in this area. Richards et al. (2000) especially those during the Last Glacial. However, detailed mapping adopted OSL dating to glacial deposits, mainly of sand lenses, in of the glacial landforms in the entire area is essential not only to which assigned past glacial advances occurred at w18 ka, w10 ka, elucidate the classification of the glacial advanced stage by mor- and w1 ka. In addition, Finkel et al. (2003) applied TCR dating to phostratigraphy but also to determine the appropriate sites at boulders in this area, and they proposed that seven glacial which to collect TCR or OSL dating samples. advanced stages have occurred 86 6 ka, 35 3 ka, 23 3 ka, 16 2 ka, 9.2 0.2 ka, 3.6 0.3 ka and 500 BP to present. 3. Methods The Kanchenjunga Himal lies in the easternmost part of the Nepal Himalayas. Existing glaciers spread over the area almost This study initially delineated the maps of glacial landforms and more than 5000 m a.s.l. and the total glacierized area reaches contemporary glaciers using the latest aerial photographs of all 314 km2 (Asahi, 1999). The Ghunsa Khola and the Shimbuwa study areas. The present ELA was identified by the surface topog- Khola valleys both originating from the peak of Kanchenjunga raphy of glacier. ELAs during the LGM were calculated by the have well-developed landforms related to the past glaciation. The maximum elevation of lateral moraines (MELMs) method after the history of these glaciations was proposed by several German compilation of the geomorphic map, stratigraphical classification of geographers. Kuhle (1990), Meiners (1999), and Ko¨nig (1999) the moraines, and relative and absolute dating control. mentioned such landforms, but their idea is based on the story of the ‘Tibetan Ice Sheet’ (Kuhle, 1988b). They proposed a drastic and 3.1. Aerial photograph interpretation large glacial extension during the Last Glaciation in Ghunsa and Shimbuwa Khola drainage that the glacier originated at the outlet Initially, vertical aerial photographs of 1:50,000 scale were of the Tibetan Ice Sheet reached an altitude of 890 m in the interpreted using a stereoscope for the entire part of the study area. The photographs were taken by the Survey Department of Nepal in 1992. More than 400 photographs were investigated during this

Fig. 2. Correlation of the past glaciations in the Khumbu and Kanchenjunga Himals in the eastern Nepal Himalayas. The bars are positioned to indicate terminus altitudes and credible time ranges of each glacial stage. Bar widths cover most of the terminus altitudes range. Symbols beside the bars denote the date of each glacial stage: circle Fig. 3. Profiles of the present and past equilibrium-line altitudes (ELA) at (A) Khumbu symbol shows interglacial or deglaciation age and triangle symbol means glacial Himal and (B) Kanchenjunga Himal. The site values of ELA are projected on the lat- advanced age. Dates in the Khumbu Himal from Richards et al. (2000) by OSL dating, itudinal plane surface. ELA of the modern glacier is quoted from Asahi (1999), which Finkel et al. (2003) by TCR dating, and Ro¨tlisberger (1986) by radiocarbon dating, and was identified from the surface topography of the contemporary glacier using the those in the Kanchenjunga Himal from Tsukamoto et al. (2002) by OSL dating and latest aerial photographs. The maximum elevation of lateral moraine (MELM) is Asahi (2004) by radiocarbon dating. The Oxygen isotopic curve quoted from Martinson interpreted as an ELA during the LGM, with the aid of the mean elevation of glacier et al. (1987). (MEG). K. Asahi / Quaternary International 212 (2010) 26–34 29 work. Existing glaciers and moraines were carefully delineated on Kanchenjunga Himals. Described moraines were divided into five the 1:50,000 scale topographic maps. Field observations, per- moraine complexes based on geomorphic stratigraphy resem- formed ahead of the aerial photograph interpretation, allowed bling that explained by Rose and Menzies (1996). In addition, accurate and more detailed interpretation of the images. The relative dating tests, mainly based on the weathering criteria of compilation of the geomorphic and glacial map was followed by boulders, were applied at 27 sites in Khumbu and 40 sites in a main field survey, and was supplemented accordingly to enable Kanchenjunga. Relative dating values verified the significant the completion of the distribution map of the moraines and difference among the five moraine complexes through the contemporary glaciers. statistical Student’s t-test (Asahi, 2004). Then the classified five moraine complexes revealed the existence of, at least, five glacial 3.2. Determination of contemporary ELA advance stages, in both areas. From younger-to-older order, the stages in the Khumbu Himal were named the Lobche II, Lobuche The present ELA of each glacier was identified from the I, Thukla, Periche II, and Periche I. In the same way, the Ramze, surface topography of the glacier. The conditions of the existing Lapsang, Yalung, Cheram, and Shimbuwa stages are found in the glacier, i.e., its shape, terminus position, range of the debris- Kanchenjunga Himal. The timing of each stage was ascertained covered area, and medium moraine, were delineated on a topo- in reference to OSL age (Richards et al., 2000) and TCR age graphical map. The maximum elevation of the debris-covered (Finkel et al., 2003) in Khumbu and to OSL age (Tsukamoto area can be an indicator of the present ELA. Along with the et al., 2002) in Kanchenjunga (Fig. 2). The recognition of glacial maximum altitude of the debris-mantle, a firn line on glacier, stage and its chronology mentioned in this article, therefore, glaciation threshold on surrounding ridge, and the glacier surface does not partly accord with those proposed by Richards et al. change from convex (ablation area) to concave (accumulation (2000) and Finkel et al. (2003). The full text of the stage clas- area) were also applied to estimate the present ELA. However, sification and history of glaciation will be described in a separate the shapes of glaciers appearing in the Himalayas are generally paper. complex because of the steep topography. Thus, only reliable In order to estimate the ELAs during the LGM, this study applied data were considered in this study, and among the existing 510 the maximum elevation of lateral moraine (MELM) and the glaciers throughout the study areas, 189 (37%) were eligible for medium elevation of the glacier (MEG) those proposed by Dahl and use in the estimation of the present ELA. Details of this selection Nesje (1992) and Nesje and Dahl (2000). Distinguished higher were explained in the Glacier Inventory of eastern Nepal (Asahi, altitudes of the MELM were selected. The MEG was also applied but 1999). only to small and shallow cirque glaciers developed only during the LGM to confirm that the reconstructed ELAs by both MELM and 3.3. Reconstruction of ELA during the LGM MEG are well correlated. On the basis of the compilation of the glacial geomorphological map covering the study area, this method Aerial photographs interpretation enabled delineation of can demonstrate the maximum amount of ELA depression, as well moraine mapping of the entire areas in the Khumbu and as paleoclimate conditions.

m a.s.l. 8000

6000

4000

2000

28°00' 28°20' 28°40' 29°00'

Fig. 4. Latitudinal variations of precipitation along the Dudh Koshi valley in the southern ﬂank of the Khumbu Himal. The histogram shows mean monthly precipitation with mean annual precipitation displayed by locality name. The data for Lobuche are cited from Tartari et al. (1998), Dingboche from ‘‘Snow and Glacier Hydrology Section Year Book 1987 to 2000’’ (nine volumes from Department of Hydrology and Meteorology (DHM), Kathmandu), and others from ‘‘Climatological Records of Nepal 1966 to 2000’’ (17 volumes from DHM). The data period for each site is; Lobuche: 1994–1996 (3 yrs); Dingboche: 1987–2000 (14 yrs); Kumjung: 1968–1991 (24 yrs); Namche Bazar: 1949–1983 (35 yrs); Chauri Kharka: 1949–2000 (52 yrs); Chialsa: 1967–1998 (32 yrs); Aiselukharka: 1949–2000 (52 yrs); Diktel: 1973–2000 (28 yrs); Kurule Ghat: 1948–2000 (53 yrs). 30 K. Asahi / Quaternary International 212 (2010) 26–34

Fig. 5. Distribution map of existing glaciers and moraines in the entire Khumbu Himal. Moraines of the LGM period were extracted using relative and numerical dating. Among these, the distinguished higher altitude of the MELM and small relative height of the MEG were selected at 13 sites in the Khumbu Himal.

4. Results valley located in the southern flank of the Great Himalayas which is originating from Mt. Everest, and were compiled in Fig. 4 to explain 4.1. Contemporary ELA the effects of the SW monsoon. Maximum precipitation (>2000 mm) occurred at the foot of the mountains, whereas scarce Contemporary ELAs, determined by surface topography of amounts of precipitation remained in the uppermost valley glacier and glaciation threshold, were plotted on a latitudinal (<500 mm in Lobuche). Enhanced insolation encourages glacier profile at Khumbu (Fig. 3A) and Kanchenjunga (Fig. 3B), respec- ablation due to the reduction of monsoon rainfall and clouds to tively. Due to the limitations of the topographical map resolution, north. Thus, ELA rises in the north, and the inclination of ELA is however, the ELA was assigned to 50 m intervals. enhanced in the latitudinal profile. The regression line in Fig. 3A shows a latitudinal profile of the The lowest elevation of the EL appeared at 5400 m in the present ELA in the Khumbu Himal. The lowest ELA that appeared in Khumbu Himal. However, the Shorong Himal is located at ca. the southernmost part of this region is presently at an altitude of ca. 15 km further to the south. ELA has been observed at AX010 5400 m, while a northward rising of the ELA gradient exceeds more glacier in Shorong and calculated at about 5200 m (Ageta and than 500 m in the north. Similarly in the Kanchenjunga Himal, ELA Higuchi, 1984). The lowest altitude of the EL, in other words, rises from south to north over altitude ranging from 5200 to appeared at 5200 m in the southern margin of Mt. Everest region, 6000 m. Conclusively, latitudinal profiles of ELA display southward that being similar to Kanchenjunga. Therefore, the lowest ELA in declination comparable to those in eastern Nepal. The same each area converges at around 5200 m at the southernmost ends southward inclination of the ELA was illustrated by Mu¨ ller (1980), of the both mountains. This altitudinal convergence in eastern Williams (1983) and Burbank and Cheng (1991) in the cross-section Nepal implies that air temperature controls ELA. Glaciers are through Mt. Everest. A northward steep gradient of the latitudinal mainly nourished by sources of summer monsoon moisture in profile of the ELA is attributed to the reduction in precipitation to this region. Monsoon vapor were brought as rainfall or snowfall the north, causing a rain shadow effect (Yasunari and Inoue, 1978; depending on the altitude, namely air temperature (Ageta and Barry, 1992). Precipitation data are available along the Dudh Koshi Kadota, 1992). Furthermore, ELAs at the southern margin are K. Asahi / Quaternary International 212 (2010) 26–34 31

Fig. 6. Photographs of ELA reconstruction sites in the Khumbu Himal. Star symbol denotes the location where MELM was applied. (A) Loc.9 (on the left) and 10 (on the right); (B) Loc.6; (C) Loc.4; (D) Loc.5; (E) 2 (on the right) and 3 (on the left). The localities are given in Fig. 5.

responsive to summer air temperature throughout the Nepal valley glaciers, such as the Khumbu Glacier (Figs. 5 and 6). These Himalayas. moraines were confirmed as those of the LGM by means of their The majority of annual accumulation nourished by the SW correlation to moraine aged LGM with the aid of relative dating monsoon and ablation occurs simultaneously in summer. Summer methods. The configuration of these moraines with the contem- precipitation and air temperature strongly control glacier mainte- porary glaciers proves that glacier extension was much limited nance. These unique and distinctive characteristics appeared only during the LGM at least in the uppermost ablation area. The in this region where the summer monsoon dominates the climatic lowering of ELA was about 1200 m in the southern end, whereas system. Ohmura et al. (1992) mentioned ELA sections in the major only 200 m in the northern. Similarly during the LGM, it showed mountains of the world. However, the steep ELA gradient was not the same inclination as that of the present, although the gradient of explained, rising almost 1000 m within 40 km in latitude. the line was much larger during the LGM. In the same way, 17 sites were selected from the Kanchenjunga 4.2. ELA during the LGM Himal (Fig. 7), and the comparison of ELA in the present and LGM is shown in Fig. 3B. The past ELAs range from 4300 to 5800 m (Table The morphostratigraphy of the glacial landforms led to the 1). The highest value of 5800 m was perhaps anomalous, but the classification into five moraine complexes, which were recognized LGM’s ELA shows the same inclination as that of the present. The in both study areas, and each glacial stage was ascertained by lowering of ELA was about 1000 m in the southernmost end and numerical age (Fig. 2). Moraines of the Periche II and Cheram stages 500 m in the northernmost. The gradient was emphasized during formed during the LGM were separated. Distinguished higher the LGM when compared with that of the present. altitudes of the MELM were selected. Regression lines were calculated from the MELM values. The In the Khumbu Himal, 13 sites were chosen to estimate past MEG values applied only to small and shallow cirque glaciers with ELAs throughout the area (Fig. 5). Fig. 6 shows the representative good correlations. Furthermore, the regional ELA is, theoretically, sites, and the descriptions of each site are listed in Table 1. The subject to a range of toe-to-summit altitude. These lines almost comparison of ELA in the present and LGM is shown in Fig. 3A. The intersect the relative height of the glacier (Fig. 3) and, consequently, altitude of 4200 m is the lowest ELA in this area. In the northern can represent the regional ELA during the LGM. Previous studies by part, higher moraine terraces occur beside the debris-covered Williams (1983) and Duncan et al. (1998) proposed ca. 1000 m 32 K. Asahi / Quaternary International 212 (2010) 26–34

Table 1 Reconstructed ELAs during the LGM for the two mountains in eastern Nepal.

Loc. Latitude (N) Longitude (E) Headwall Terminus Relative Mean elevation of Maximum elevation

0 000 00altitude (m) altitude (m) height (m) glacier (MEG) (m) of lateral moraine (MELM) (m) Khumbu Himal 1 275955865030 5480 2 275915864805 5380 3 275815864855 5270 4 275530865145 5200 5 275515865350 5225 6 275410865010 5370 7 27 53 0 86 45 00 5360 4960 400 5160 5000 8 275245865550 5000 9 275055864835 4440 10 27 50 40 86 48 30 4949 4160 789 4555 4560 11 27 50 20 86 48 15 4949 4180 769 4565 4440 12 27 50 0 86 51 15 5180 13 27 44 20 86 40 40 4400 3500 900 3950 4200

Kanchenjunga Himal 1 275040880510 5800 2 274840880250 5250 3 27 48 10 88 02 30 5460 5040 420 5250 5260 4 27 47 50 88 02 30 5360 4960 400 5160 5200 5 274650880100 4880 6 274240875720 4800 7 27 39 00 87 50 50 4772 4000 772 4386 4370 8 27 38 50 87 51 10 4560 4200 360 4380 4320 9 27 38 30 87 50 00 4346 4160 186 4253 4300 10 27 37 30 87 53 50 4560 4280 280 4420 4390 11 27 37 00 87 57 20 4580 12 27 36 00 87 56 50 4560 13 27 34 50 87 56 00 4500 4240 260 4370 4300 14 27 34 30 87 56 20 4787 4360 427 4574 4480 15 27 34 10 87 56 00 4636 4440 196 4538 4490 16 27 33 30 88 00 00 4490 17 27 32 20 87 57 40 4340

The locality where MELM and MEG were applied should be referred to Fig. 5 for the Khumbu Himal and to Fig. 7 for the Kanchenjunga Himal, respectively. depression from Mt. Everest region and the Narayani river system, north played a great role on the characteristics of glaciers in this respectively. The general tendency of this depression from both region. This means that monsoon precipitation was a dominant Khumbu and Kanchenjunga agrees with that in the previous factor on glacier accumulation during the LGM period as well. studies in the south margin of the glaciated area. Meanwhile, the upwelling in the Arabian Sea was at a minimum Owen and Benn (2005) reconstructed the ELAs during the LGM phase around 18 ka (van Campo et al., 1982; Prell and Kutzbach, in the uppermost of the Khumbu–Imja confluence valley. They 1992; Sirocko, 1996). As a consequence of weakened upwelling, inferred glacier extension from preserved moraines and upper limit Ono et al. (2004) stated that a weakening and even a possible loss of of ice-scoured bedrocks, and recognized that large medial moraine the SW summer monsoon during a part of the Last Glacial, espe- implies a minimum elevation of the LGM glacier ELAs. They cially during the LGM. In case of SW monsoon extinction, during concluded that 5400 m is the representative value in this area, and the LGM, winter rainfall by westerlies should be a ruling factor on calculated the lowering altitude of ELA was 200–300 m, comparing glacier formation throughout the Nepal Himalayas. These glaciers with the contemporary ELA in the upper Khumbu area (Asahi, might be situated where the climatic condition was almost the 1999). Reconstructed ELA in this study and above estimates have same as those in the Karakoram and Hindukush mountains at close concordance. Site specific values of former ELAs are insuffi- present. In these mountains, the longitudinal profile of ELA declines cient in Owen and Benn (2005); however, this study confirmed the westward due to the effect of winter westerlies bringing moisture validity of the past ELAs with field evidence throughout the area. from the Mediterranean, Black, and Caspian Seas (Wissmann, 1959; Holmes, 1993). However, the lowest ELAs at the southern ends 5. Paleo-monsoon implications during the LGM converged at altitudes of approximately 4200 m in Khumbu and Kanchenjunga (Fig. 3). The same tendency in the Inclinations of latitudinal ELAs both of the present and the LGM longitudinal convergence and latitudinal gradient between the are suggestive. In the Khumbu Himal, ELA rises northward at a rate LGM and the present suggests that the climate during the LGM was of 8.3 m km1 for 40 km in the southern flank of the Great under similar climatic conditions as those in the present. Himalayas. Moreover, the gradient during the LGM was During the LGM period, the latitudinal ELA profiles show the 42.4 m km1. Conspicuous latitudinal contrast in eastern Nepal same southward inclination tendency, suggesting that monsoon could not be explained only by the gradient of temperature, but precipitation is likely to be a ruling factor on the glacier accumu- indicates that monsoon precipitation is likely to be a ruling factor on lation. Hence, the Nepal Himalayas were in a summer monsoon glacier nourishment. Dominant precipitation by the southwestern environment during the LGM as present, even if the SW monsoon monsoon causes the sharp declination of precipitation in the was weakened. Simulations demonstrated arid conditions and less southern slope of the Himalayas (Fig. 4). Therefore, during the LGM, precipitation during the LGM in South Asia (Prell and Kutzbach, the same southward inclinations of the ELAs at present (Fig. 3) show 1987). Indeed, glacial advances were surely constrained in the that the remarkable contrasts of precipitation between south and Nepal Himalayas by monsoon precipitation and cooling air K. Asahi / Quaternary International 212 (2010) 26–34 33

Fig. 7. Distribution map of existing glaciers and moraines in the entire Kanchenjunga Himal, same as Fig. 5, but in Kanchenjunga for 17 MELM and MEG selection sites. (A) moraine of the Cheram stage (LGM); (B) moraine of other stages; (C) site for MELA and/or MEG applied; (D) existing glaciers (bare ice); (E) existing glaciers with debris-mantled.

temperature. Therefore, the SW monsoon might have been weaker Ageta, Y., Higuchi, K., 1984. Estimation of mass balance components of a summer- during the LGM, but its existence was certainly significant. accumulation type glacier in the Nepal Himalaya. Geografiska Annaler 66A, 249–255. Ageta, Y., Kadota, T., 1992. Predictions of changes of glacier mass balance in the Acknowledgements Nepal Himalaya and Tibetan Plateau: a case study of air temperature increase for three glaciers. Annals of Glaciology 16, 89–94. Asahi, K., 1999. Data on inventoried glaciers and its distribution in eastern part of The major part of this study was carried out while I was in Nepal Himalaya. In: Data Report 2, Basic Studies for Assessing the Impacts of Tribhuvan University, Kathmandu as a research fellow under the the Global Warming on the Himalayan Cryosphere, 1994–1998. Institute for Asian Studies Program from the Ministry of Education, Science and Hydrosperic-Atmospheric Sciences, Nagoya University and Department of Hydrology and Meteorology, Nepal, pp. 1–76. Culture, Japan. I would like to extend my gratitude to T. Watanabe of Asahi, K., 2004. Late Pleistocene and Holocene glaciations in the Nepal Himalayas Hokkaido University and anonymous reviewers for improvement of and their implications for reconstruction of paleoclimate. PhD thesis, Hokkaido the manuscript. This study was partly supported by Grant-in-Aid University, 127p. for Scientific Research from the Japan Society for the Promotion of Asahi, K., Watanabe, T., 2000. Past and recent glacier fluctuations in Kanchenjunga Himal, Nepal. Journal of Nepal Geological Society 22, 481–490. Science (No. 14000268) and the Sasagawa Grants for Science Barnard, P.L., Owen, L.A., Finkel, R., Asahi, K., 2006. Landscape response to degla- Fellows by the Japan Science Society. ciation in a high relief, monsoon-influenced alpine environment, Langtang Himal, Nepal. Quaternary Science Reviews 25, 1391–1411. Barry, R.G., 1992. Mountain Weather and Climate, second ed. Routledge, London. References Benn, D.I., Lehmkuhl, F., 2000. Mass balance and equilibrium-line altitudes of glaciers in high-mountain environments. Quaternary International 65/66, 15–29. Abramowski, U., Glaser, B., Kharki, K., Kubik, P.W., 2006. Late Pleistocene and Benn, D.I., Owen, L.A., 1998. The role of the Indian summer monsoon and the mid- Holocene paleoglaciations of the Nepal Himalaya: 10Be surface exposure dating. latitude westerlies in Himalayan glaciation: review and speculative discussion. Zeitschrift fu¨ r Gletscherkunde und Glazialgeologie 39, 183–196. Journal of Geological Society 155, 353–363. 34 K. Asahi / Quaternary International 212 (2010) 26–34

Benn, D.I., Owen, L.A., Osmaston, H.A., Seltzer, G.O., Porter, S.C., Mark, B., 2005. Ono, Y., Shulmeister, J., Lehmkuhl, F., Asahi, K., Aoki, T., 2004. Timings and Reconstruction of equilibrium-line altitudes for tropical and sub-tropical causes of glacial advances across the PEP-II transect (East Asia to glaciers. Quaternary International 138/139, 8–21. Antarctica) during the last glaciation cycle. Quaternary International 118/ Burbank, D.W., Cheng, K.J., 1991. Relative dating of Quaternary moraines, Rongbuk 119, 55–68. Valley, Mount Everest, Tibet: implications for an ice sheet on the Tibetan Owen, L.A., Benn, D.I., 2005. Equilibrium-line altitudes of the Last Glacial Maximum Plateau. Quaternary Research 36, 1–18. for the Himalaya and Tibet: an assessment and evaluation of results. Quaternary Burbank, D.W., Blythe, A.E., Putkonen, J., Pratt-Sltaula, B., Gabet, E., Oskin, M., International 138/139, 55–78. Barros, A., Ojha, T.P., 2003. Decoupling of erosion and precipitation in the Owen, L.A., Mitchell, W.A., Bailey, R.M., Coxon, P., Rhodes, E.J., 1997. Style and timing Himalayas. Nature 426, 652–655. of glaciation in the Lahul Himalaya, northern India: a framework for recon- van Campo, E., Duplessy, J.C., Rossignol-strick, M., 1982. Climatic conditions structing Late Quaternary paleoclimatic change in the western Himalayas. deduced from a 150-kyr oxygen isotope-pollen record from the Arabian Sea. Journal of Quaternary Science 12, 83–109. Nature 296, 56–69. Owen, L.A., Derbyshire, E.,, Fort, M., 1998. The Quaternary glacial history of the Dahl, S.O., Nesje, A., 1992. Paleoclimatic implications based on equilibrium-line Himalaya. Quaternary Proceedings 6, 91–120. altitude depressions of reconstructed Younger Dryas and Holocene cirque Pant, G.B., Kumar, K.R., 1997. Climates of South Asia. John Wiley & Sons, West glaciers in inner Nordfjord, western Norway. Palaeogeography Palae- Sussex. oclimatology Palaeoecology 94, 87–97. Porter, S.C., 2001. Snowline depression in the tropics during the Last Glaciation. Duncan, C.C., Dlein, A.J., Masek, J.G., Idsvkd, B.L., 1998. Comparison of late Pleisto- Quaternary Science Reviews 20, 1067–1091. cene and modern glacier extents in central Nepal based on digital elevation Prell, W.L., Kutzbach, J.F., 1987. Monsoon variability over the past 150,000 years. data and satellite imagery. Quaternary Research 49, 241–254. Journal of Geophysical Research 92, 8411–8425. Finkel, R.C., Owen, L.A., Barnard, P.L., Caffee, M.W., 2003. Beryllium-10 dating of Prell, W.L., Kutzbach, J.E., 1992. Sensitivity of the Indian Monsoon to forcing Mount Everest moraines indicates a strong monsoon influence and glacial parameters and implications for its evolution. Nature 360, 647–652. synchronicity throughout the Himalaya. Geology 31, 561–564. Richards, B.M.W., 2000. Luminescence dating of Quaternary sediments in the Fushimi, H., 1978. Glaciations in the Khumbu Himal (2). Seppyo 40 (Special issue), Himalaya and High Asia: a practical guide to its use and limitations. Quaternary 71–77. International 65/66, 49–61. Gayer, E., Lave´, J., Rik, R., France-Lanord, C., 2006. Monsoonal forcing of Holocene Richards, B.M.W., Benn, D.I., Owen, L.A., Rhodes, E.J., Spencer, J.Q., 2000. Timing of glacier fluctuations in Ganesh Himal (Central Nepal) constrained by cosmogenic late Quaternary glaciations south of Mount Everest in the Khumbu Himal, 3He exposure ages of garnets. Earth and Planetary Science Letters 252, 275–288. Nepal. Geological Society of America Bulletin 112, 1621–1632. Holmes, J.A., 1993. Present and past patterns of glaciation in the Northwest Rose, J., Menzies, J., 1996. Glacial stratigraphy. In: Menzies, J. (Ed.), Past Glacial Himalaya: climatic, tectonic and topographic controls. In: Shroder, J.F. (Ed.), Environments: Sediments, Forms and Techniques. Butterworth Heinemann, Himalaya to the Sea. Routledge, London, pp. 72–90. Oxford, pp. 253–284. Iwata, S., 1976. Late Pleistocene and Holocene moraines in the Sagarmatha (Everest) Ro¨tlisberger, F., 1986. 10000 Jahre Gletschergeschichte der Erde. Verlag Sauerlander, region, Khumbu Himal. Seppyo 38 (Special issue), 109–114. Arau. Ko¨nig, O., 1999. Preliminary results on the last-high-glacial glaciation of the Sirocko, F., 1996. The evolution of the monsoon climate over the Arabian Sea during Rolwaling Himal and the Kanchenjunga Himal (Nepal, E-Himalaya). Geojournal the last 25,000 years. Palaeoecology of Africa 24, 53–69. 47, 373–384. Shiraiwa, T., 1993. Glacial fluctuations and cryogenic environments in the Langtang Kuhle, M., 1982. Der Dhaulagiri- und Annapurna-Himalaya. Zeitschrift fu¨ r Valley, Nepal Himalaya. Contributions from Low Temperature Science Hokkaido Geomorphologie, N.F. 41, 1–229. Univ Ser.A 38, 1–98. Kuhle, M., 1988a. Topography as a fundamental element of glacial systems – a new Tartari, G., Verza, G., Bertolami, L., 1998. Meteorological data at the Pyramid approach to ELA calculation and typological classification of paleo- and recent observatory laboratory (Khumbu Valley, Sagarmatha National Park, Nepal). glaciations. Geojournal 17, 545–568. Memorie. Dell’Istituto Italiano di Idrobiologia 57, 23–40. Kuhle, M., 1988b. Geomorphological findings on the build-up of Pleistocene glacia- Tsukamoto, S., Asahi, K., Watanabe, T., Rink, W.J., 2002. Timing of past glaciations in tion in southern Tibet, and on the problem of inland ice. Geojournal 17, 457–511. Kanchenjunga Himal, Nepal by optically stimulated luminescence dating of tills. Kuhle, M., 1990. New data on the Pleistocene glacial cover of the southern border of Quaternary International 97/98, 57–67. Tibet: the glaciation of the Kangchendzo¨nga Massif (8585 m E-Himalaya). Williams, V., 1983. Recent and former equilibrium-line altitudes near Mount Geojournal 20, 319–323. Everest, Nepal and Tibet. Arctic and Alpine Research 15, 201–211. Martinson, D.G., Pisias, N.G., Hays, J.D., Imbrie, J., Moore Jr., T.C., Shackleton, N.J., Wissmann, H. von, 1959. Die heutige Vergletscherung und Schneegrenze in 1987. Age dating and the orbital theory of the Ice Age: development of a high- Hochasien, mit Hinweisen auf die Vergletscherung der letzten Eiszeit. Akade- resolution 0 to 300,000-year chronostratigraphy. Quaternary Research 27, 1–29. mie der Wissenschaften und der Literatur in mainz. Abhandlungen der Meiners, S., 1999. The history of glaciation of the Rolwaling and Kanchenjunga Mathematisch-Naturwissenschftlichen Klasse 14, 1101–1431. Himalaya. Geojournal 50, 341–372. Yasunari, T., Inoue, J., 1978. Characteristics of monsoon precipitation around Mu¨ ller, F., 1980. Present and late Pleistocene equilibrium line altitudes in the Mt peaks and ridges in Shorong and Khumbu Himal. Seppyo 40 (Special issue), Everest region – an application of the glacier inventory. International Associa- 26–32. tion of Hydrological Science Publication 126, 75–94. Zech, W., Glaser, B., Abramowski, U., Dittmar, C., Kubik, P.W., 2003. Reconstruction Nesje, A., Dahl, S.O., 2000. Glaciers and Environmental Change. Arnold, London. of the Late Quaternary glaciation of the Macha Khola valley (Gorkha Himal, Ohmura, A., Kasser, P., Funk, M., 1992. Climate at the equilibrium line of glaciers. Nepal) using relative and absolute (14C, 10Be, dendrochlonology) dating Journal of Glaciology 38, 397–411. techniques. Quaternary Science Reviews 22, 2253–2265.